摘要
针对一类未知非线性系统,设计了一种基于小波神经网络的自适应控制器,并提出了一种适合在线学习的参数混合训练算法。根据离线和在线学习系统的特性,得到小波神经网络控制器的初始参数,使用混合训练算法在线修正控制律,实现了自适应控制。仿真结果验证了该控制方案的有效性。
An adaptive controller based on wavelet neural network was designed for a class of unknown nonlinear system. And a mixed training algorithm suitable for online parameter adjusting was presented. By learning offline and online characteristics of the system, the initial parameters of wavelet neural network controller were obtained. The control law online for adaptive control can be modified by the mixed training algorithm. The simulation results demonstrate the validity of proposed controller.
出处
《中国石油大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2007年第5期141-143,147,共4页
Journal of China University of Petroleum(Edition of Natural Science)
基金
国家'973'项目(2004CB31800)
关键词
小波神经网络
自适应控制器
在线学习算法
wavelet neural network
adaptive controller
online learning algorithm
